TWI804293B - Handheld physiological monitoring device - Google Patents

Abstract

A handheld physiological monitoring device includes a casing, a first temperature sensor, a second temperature sensor, a humidity sensor, a display unit, and a control unit. When a user holds a grip portion of the casing with one hand and the control unit receives a trigger signal, the control unit controls the second temperature sensor to measure a plurality of palm temperatures and reads a plurality of ambient temperatures measured by the first temperature sensor and reads a plurality of ambient humidity measured by the humidity sensor continuously with a sampling time difference. The control unit inputs a plurality of the palm temperature, the ambient temperature, and the ambient humidity into a neural network model to obtain an estimated body temperature of the user, and controls the display unit to display the estimated body temperature.

Description

Handheld Physiological Monitoring Device

The present invention relates to a physiological monitoring device, in particular to a handheld physiological monitoring device which is easy to operate and simultaneously detects multiple physiological data.

With the advancement of technology and people's emphasis on health, various wearable devices for monitoring physiological signals are also booming. Among them, wearable devices such as smart watches or similar wristbands can detect and obtain the wearer's heart rate by using photoplethysmography (PPG) technology to detect the position of the wrist. and blood oxygen concentration (SPO2). However, it is not easy to set and operate the electronic device of this smart watch for elderly users. In addition, the existing smart watches cannot provide the function of measuring the body temperature of the wearer, or cannot provide relatively accurate body temperature measurement results. Therefore, whether there are other physiological measurement or monitoring devices capable of simple operation and body temperature measurement, and then integrated detection of other existing physiological signals has become a problem to be solved.

Therefore, the object of the present invention is to provide a handheld physiological monitoring device that is easy to operate.

Therefore, the present invention provides a handheld physiological monitoring device, which is suitable for a user and includes a casing, a first temperature sensor, a second temperature sensor, a humidity sensor, and a display unit , and a control unit.

The housing includes a grip portion adapted to be held by the user's palm. The first temperature sensor is disposed on the housing and used for measuring an ambient temperature. The second temperature sensor is arranged on the holding part and is used for measuring a palm temperature of the palm. The humidity sensor is arranged on the casing and used for measuring an environment humidity. The display unit is arranged on the casing and used for displaying information.

The control unit is arranged in the housing and electrically connected to the first temperature sensor, the second temperature sensor, the humidity sensor, and the display unit, and stores a type of neural network model. When the user's palm grasps the holding part and the control unit receives a trigger signal, the control unit reads the temperature measured by the second temperature sensor multiple times continuously with a sampling time difference. palm temperature, the ambient temperature measured by the first temperature sensor, and the ambient humidity measured by the humidity sensor, and the palm temperature, the ambient temperature, and the The ambient humidity is input into the neural network model to obtain an estimated body temperature of the user, and the display unit is controlled to display the estimated body temperature.

In some implementation aspects, the handheld physiological monitoring device also includes an electrical connection An input unit connected to the control unit, wherein the input unit is suitable for the user to operate to generate the trigger signal.

In some implementation aspects, the gripping portion includes a curved surface located above, and a second opening formed on the curved surface. The curved surface is suitable for sticking to the palm. The second temperature sensor is a sensing element used to measure body temperature and convert it into an electrical signal, and is disposed adjacent to the second opening.

In some implementation aspects, the housing further includes a base portion extending from below the handle portion, the base portion includes a first opening, the first temperature sensor and the humidity sensor The device is disposed adjacent to the first opening, so that when the palm adheres to the curved surface of the holding portion, the palm will not cover the first opening.

In some embodiments, the base part further includes a bottom surface and a third opening formed on the bottom surface. The input unit is an input button arranged on the bottom surface, and the display unit is a screen arranged in the third opening.

In some implementation aspects, the handheld physiological monitoring device further includes an optical volume sensor electrically connected to the control unit and disposed adjacent to the second opening. Wherein, when the palm of the user grasps the grip part and the control unit receives the trigger signal from the input unit, the control unit controls the optical volume sensor to emit a red incident light and an infrared ray incident light to the palm, and receive corresponding two reflected lights to generate a first light volume signal and a second light volume signal, or control the light volume sensor to emit a green incident light to the palm, and receive corresponding to another reflected light to produce a third photovolume signal. The control unit calculates a heart rate, a blood pressure, and a blood oxygen concentration of the user according to the first light volume signal and the second light volume signal, or, according to the first light volume signal and the second light volume signal The blood oxygen concentration of the user is calculated and the heart rate and blood pressure are calculated according to the third photovolume signal.

In some implementation aspects, the handheld physiological monitoring device further includes a communication unit electrically connected to the control unit, wherein the communication unit supports wireless communication transmission technology, after the control unit obtains the estimated body temperature, by The communication unit transmits the estimated body temperature, the heart rate, and the blood oxygen concentration to an electronic device of the user.

In some other implementation aspects, wherein the neural network model is a supervised machine learning model, and the palm temperatures, the ambient temperatures, the ambient humidity, and multiple actual body temperature for training.

The effect of the present invention is that as long as the user grasps the grip part with the palm and the control unit receives the trigger signal, the control unit can input the palm temperatures, the ambient temperatures, and the ambient humidity The neural network model obtains the estimated body temperature of the user, and displays the estimated body temperature through the display unit, so that the user can know the measurement result immediately.

1: Control unit

2: The first temperature sensor

3: Second temperature sensor

4: Humidity sensor

5: Display unit

6: Input unit

7: Optical volume sensor

8: Communication unit

9: Housing

91: Grip

92: base part

93: First opening

94: Second opening

95: The third opening

96: Charging opening

97: Input opening

Other features and effects of the present invention will be clearly presented in the implementation manner with reference to the drawings, wherein: Fig. 1 is a block diagram illustrating an embodiment of the handheld physiological monitoring device of the present invention; Fig. 2 is a perspective view illustrating a housing of the embodiment; Fig. 3 is a perspective view assisting Fig. 2 in illustrating the embodiment A base portion of the housing; and FIG. 4 is a schematic diagram illustrating the structure of a neural network model of the embodiment.

Before the present invention is described in detail, it should be noted that in the following description, similar elements are denoted by the same numerals.

Referring to Fig. 1, one embodiment of the handheld physiological monitoring device of the present invention is suitable for a user and includes a housing 9, a first temperature sensor 2, a second temperature sensor 3, a humidity A sensor 4 , a display unit 5 , an input unit 6 , an optical volume sensor 7 , a communication unit 8 , and a control unit 1 .

Referring to FIG. 1 , FIG. 2 , and FIG. 3 , the housing 9 includes a handle portion 91 and a base portion 92 extending below the handle portion 91 . The holding portion 91 is suitable for being held by the user's palm. In this embodiment, the grip portion 91 is similar to a sphere and includes a curved surface on the top, and a second opening 94 formed on the curved surface, the curved surface is suitable for the palm to be attached to hold the grip portion 91 . The base portion 92 includes a first opening 93, a charging opening 96, a bottom surface, and a third opening 95 formed on the bottom surface and an input Hole 97.

The first temperature sensor 2 is, for example, a thermistor, and is disposed on the casing 9 and adjacent to the first opening 93 for measuring an ambient temperature. The second temperature sensor 3 is, for example, an infrared temperature sensor, a thermistor, or a thermopile, which can measure body temperature, and is arranged on the grip portion 91 and adjacent to the second opening 94 for use in for measuring the palm temperature of the palm. The humidity sensor 4 is, for example, a resistive or capacitive humidity sensor, and is disposed on the casing 9 and adjacent to the first opening 93 for measuring an ambient humidity. The charging opening 96 is suitable for setting a charging connector to provide a cable (not shown) to be plugged in to charge the handheld physiological monitoring device. The charging connector is, for example, a Universal Serial Bus (USB) Type-C connector or Micro-USB connector. In this embodiment, the first opening 93 is located on the side of the base portion 92 , so that when the palm attaches to the curved surface of the gripping portion 91 , the palm will not cover the first opening 93 . In other embodiments, the first opening 93 may also be disposed at other positions of the base portion 92 (such as around the charging opening 96 ).

The input unit 6 is suitable for the user to operate to generate a trigger signal. In this embodiment, the input unit 6 is an input button disposed on the bottom surface, and is disposed adjacent to the input opening 97 so that the user can press the input button through the input opening 97 . The display unit 5 is arranged on the housing 9 and adjacent to the third opening 95, and is, for example, a screen and related driving circuits, so that the user can view the display unit 5 through the third opening 95. Information. The screen is, for example, a Organic Light Emitting Diode (OLED) screen, or Liquid Crystal Screen (LCD), etc. The communication unit 8 supports wireless communication transmission technology, such as Bluetooth transmission protocol, but not limited thereto.

The optical volume sensor 7 is disposed in the casing 9 and adjacent to the second opening 94, and is used to emit a red incident light, an infrared incident light, or a green incident light, and receive corresponding reflected light to adopt light A first photovolume signal corresponding to red light, a second photovolume signal corresponding to infrared light, or a third photovolume signal corresponding to green light is generated by the plethysmography method.

The control unit 1 is, for example, a microcontroller, a microprocessor, or a digital signal processor, etc., and is arranged in the casing 9, and is electrically connected to the first temperature sensor 2, the second temperature sensor Detector 3, the humidity sensor 4, the display unit 5, the input unit 6, the light volume sensor 7, and the communication unit 8, and store a type of neural network model. This type of neural network model is a supervised machine learning model, which is trained in advance with known and corresponding multiple palm temperatures, multiple ambient temperatures, multiple ambient humidity, and multiple actual body temperatures. Each corresponding actual body temperature is obtained, for example, by measuring the user with an ear thermometer.

In more detail, because the body temperature of the human body is a constant temperature, that is, the body temperature will not change too much in a short period of time, so this type of neural network model is a time series network with an algorithm of memory function , For example: Recurrent Neural Network (RNN), Long Short-Term Memory (Long Short-Term Memory, LSTM), Gated Recurrent Unit (Gated Recurrent Unit, GRU). Referring to Figure 4 again, Figure 4 illustrates the architecture of this type of neural network model, where x is the time step (Timestep), y _s is the number of neurons in the time series layer, and y _F is the number of neurons in the fully connected layer The number, n is the number of layers of the time series layer, and m is the number of layers of the fully connected layer. The known multi-stroke (ie, time step) features are used as the input of this type of neural network model, and each feature includes a palm temperature, an ambient temperature, and an ambient humidity corresponding to the same time point. For example, the number of these time steps is 5 strokes, and the five strokes feature in sequence five groups of time 0, time t, time 2t, time 3t, and time 4t [the palm temperature, the ambient temperature, the ambient humidity ], such as [[35,28,42%], [35.1,28,42%], [35.1,28,42%], [35.1,28,41%], [35,28.1,41%]], And the corresponding actual body temperature is 36.5 degrees, t is a sampling time difference, and the range is between several milliseconds to seconds, such as 10 milliseconds. The time steps, the number of neurons in the time series layer and the number of layers can be adjusted and trained according to the performance of the device. Then, training and convergence are performed through multiple layers of fully connected layers. The number and layers of neurons in these fully connected layers are also adjusted and trained according to the performance of the device, and this type of neural network model is trained.

When the user wants to use the handheld physiological monitoring device, the user's palm holds the grip portion 91, and the other hand operates the input unit 6 (such as pressing the input button), so that The control unit 1 receives the trigger signal, and then the control unit 1 controls to read or directly read the palm temperatures measured by the second temperature sensor 3 multiple times with a sampling time difference. The temperature measured by a temperature sensor 2 The ambient temperature, and the ambient humidity measured by the humidity sensor 4, and input the palm temperature, the ambient temperature, and the ambient humidity into the neural network model to obtain the user An estimated body temperature of . The estimated body temperature is equivalent to the user's actual body temperature measured by the ear thermometer.

Continuing from the previous example, when the control unit 1 receives the trigger signal, the control unit 1 controls the second temperature sensor 3 to take five consecutive measurements with the sampling time difference to first obtain the corresponding five temperature values. And when the control unit 1 receives the trigger signal, the control unit 1 also reads the five ambient temperatures measured by the first temperature sensor 2 consecutively five times with the sampling time difference, and the humidity sensor 4 The ambient humidity measured in Wubi, and then generate five groups [the palm temperature, the ambient temperature, and the ambient humidity] at time 0, time t, time 2t, time 3t, and time 4t as the data sequence input into the model .

In addition, when the control unit 1 receives the trigger signal, the control unit 1 also controls the light volume sensor 7 to emit the red incident light and the infrared incident light to the palm, and receive the corresponding two reflected lights to The first light volume signal and the second light volume signal are generated. The control unit 1 calculates a heart rate, a blood pressure (such as systolic blood pressure and diastolic blood pressure), and a blood oxygen concentration of the user according to the first light volume signal and the second light volume signal. Alternatively, the control unit 1 also controls the light volume sensor 7 to emit the green incident light to the palm, and receive another corresponding reflected light to generate the third volume signal, so that the control unit 1 can generate the third volume signal according to the first light. volume signal and the second light The blood oxygen concentration of the user is calculated from the volume signal and the heart rate and blood pressure are calculated according to the third light volume signal.

After the control unit 1 obtains the estimated body temperature, the heart rate, the blood pressure, and the blood oxygen concentration, in addition to controlling the display unit 5 to display the estimated body temperature, the heart rate, the blood pressure, and the blood oxygen concentration, it can also use A connection is established between the communication unit 8 and an electronic device of the user, and the estimated body temperature, the heart rate, the blood pressure, and the blood oxygen concentration are sent to the electronic device. For example, the electronic device is a smart phone, and a corresponding application program (Application Program, APP) has been installed, and can display or store various measurement results received.

To sum up, the user only needs to grasp the grip portion 91 with the palm of his hand, and operate the input unit 6 to trigger the trigger signal, and then he can learn all kinds of information on the display unit 5 through the known technology. The heart rate, the blood pressure, and the blood oxygen concentration calculated by the algorithm can also obtain the estimated body temperature that cannot be calculated by conventional techniques, and obtain multiple physiological measurement results together, so it can indeed achieve the purpose of the present invention Purpose.

But the above-mentioned ones are only embodiments of the present invention, and should not limit the scope of the present invention. All simple equivalent changes and modifications made according to the patent scope of the present invention and the content of the patent specification are still within the scope of the present invention. Within the scope covered by the patent of the present invention.

1: Control unit

2: The first temperature sensor

3: Second temperature sensor

4: Humidity sensor

5: Display unit

6: Input unit

7: Optical volume sensor

8: Communication unit

Claims (8)

A handheld physiological monitoring device is suitable for a user and includes: a housing including a grip adapted to be grasped by a palm of the user; A first temperature sensor is arranged on the housing and used to measure an ambient temperature; A second temperature sensor is arranged on the grip part and is used to measure a palm temperature of the palm; A humidity sensor is arranged on the casing and used to measure an environment humidity; A display unit is arranged on the housing and used for displaying information; and A control unit, which is arranged in the housing and electrically connected to the first temperature sensor, the second temperature sensor, the humidity sensor, and the display unit, and stores a type of neural network model, when used When the palm of the user grasps the holding part, and the control unit receives a trigger signal, the control unit reads the palm temperatures measured by the second temperature sensor multiple times consecutively with a sampling time difference , the ambient temperatures measured by the first temperature sensor, and the ambient humidity measured by the humidity sensor, and the palm temperatures, the ambient temperatures, and the ambient humidity are input The neural network model obtains an estimated body temperature of the user, and controls the display unit to display the estimated body temperature. The handheld physiological monitoring device according to claim 1 further includes an input unit electrically connected to the control unit, wherein the input unit is suitable for the user to operate to generate the trigger signal. The handheld physiological monitoring device as described in claim 2, wherein the grip part includes a curved surface on the top and a second opening formed on the curved surface, the curved surface is suitable for attaching to the palm, and the first The second temperature sensor is a sensing element for measuring body temperature and converting it into an electrical signal, and is disposed adjacent to the second opening. The handheld physiological monitoring device according to claim 3, wherein the housing further includes a base part extending from the grip part below, the base part includes a first opening, and the first temperature sensor The detector and the humidity sensor are disposed adjacent to the first opening, so that when the palm is attached to the curved surface of the grip portion, the palm will not cover the first opening. The handheld physiological monitoring device according to claim 4, wherein the base part further includes a bottom surface and a third opening formed on the bottom surface, and the input unit is an input button arranged on the bottom surface, The display unit is a screen arranged in the third opening. The handheld physiological monitoring device according to claim 5, further comprising an optical volume sensor electrically connected to the control unit and disposed adjacent to the second opening, wherein, when the user's palm grasps the When the control unit receives the trigger signal from the input unit, the control unit controls the optical volume sensor to emit a red incident light and an infrared incident light to the palm, and receive the corresponding two Reflect light to generate a first light volume signal and a second light volume signal, or control the light volume sensor to emit a green incident light to the palm, and receive another corresponding reflected light to generate a third light Volume signal, the control unit calculates a heart rate, a blood pressure, and a blood oxygen concentration of the user according to the first light volume signal and the second light volume signal, or, according to the first light volume signal and the second light volume signal The light volume signal calculates the blood oxygen concentration of the user and calculates the heart rate and the blood pressure according to the third light volume signal. The handheld physiological monitoring device as described in claim 6, further includes a communication unit electrically connected to the control unit, wherein the communication unit supports wireless communication transmission technology, and after the control unit obtains the estimated body temperature, it uses the communication The unit transmits the estimated body temperature, the heart rate, the blood pressure, and the blood oxygen concentration to an electronic device of the user. The handheld physiological monitoring device as described in claim 1, wherein the neural network model is a supervised machine learning model, and is known and corresponding to the palm temperature, the ambient temperature, the Wait for the ambient humidity and multiple actual body temperatures for training.

Images